Systematic, controlled therapy research on generalized anxiety disorder, a prevalent disorder characterized by chronic worry, has only recently begun. Review of extant theoretical and empirical literature including recent basic and therapy outcome research suggests that, although a combined cognitive behavioral intervention package (applied relaxation, self-control desensitization, and cognitive therapy) which targets intrapersonal anxious process is effective in treating this disorder, losses in therapeutic gains by long-term follow-up may be due to a failure of CBT to address significant interpersonal/emotional processing problems. Consequently, a CBT that includes interpersonal/ emotional processing therapy techniques is likely to yield greater and more long-lasting improvements. The proposed study will therefore compare (a) the current, best available CBT package plus a supportive listening therapy element and (b) the same CBT package plus an interpersonal/ emotional processing therapy element. The supportive listening element in the former condition is designed to equate the two conditions for total amount of therapy time. Outcome will be assessed at posttherapy and again at 6, 12, and 24 months after therapy. Several process measures and secondary outcome measures (including a psychophysiological laboratory session measuring central and peripheral physiological activity during theoretically relevant tasks and administered at pretherapy and posttherapy) and the assessment of matched nonanxious control subjects will provide information regarding the nature of the psychopathology and the nature of the mechanisms of therapeutic change.